Efficient Time-Stepping for Numerical Integration Using Reinforcement Learning

DOI10.1137/21m1412682zbMath1520.65050arXiv2104.03562MaRDI QIDQ6039265

Eyke Hüllermeier, Michael Dellnitz, Sina Ober-Blöbaum, Christian Offen, Sebastian Peitz, Karlson Pfannschmidt, Marvin Lücke

Publication date: 4 May 2023

Published in: SIAM Journal on Scientific Computing (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/2104.03562

zbMATH Keywords

initial value problems; reinforcement learning; quadrature; machine learning; time-stepping

Mathematics Subject Classification ID

68T05: Learning and adaptive systems in artificial intelligence

34A12: Initial value problems, existence, uniqueness, continuous dependence and continuation of solutions to ordinary differential equations

65L05: Numerical methods for initial value problems involving ordinary differential equations

65L06: Multistep, Runge-Kutta and extrapolation methods for ordinary differential equations

65D32: Numerical quadrature and cubature formulas